Predictive Artificial Neural Networks as Applied Tools in the Remediation of Dyes by Adsorption—A Review

Abstract

Environmental pollution has become one of society’s main concerns, which is why it is increasingly necessary to look for new effective remediation strategies. The applicability of artificial intelligence (AI) tools in different areas of knowledge has attracted the attention of the academic community due to their ability to process large datasets efficiently and deliver rapid, reliable predictions. Among them, artificial neural networks (ANNs) have stood out as some of the main aids in remediation and environmental control research. In this review, we systematically assess the current state of research on ANN applications in environmental remediation, particularly focusing on the remediation of dyes in water and wastewater treatment, through a comprehensive bibliographical analysis using the Methodi Ordinatio 2.0 methodology. This advanced methodology adds value compared to traditional systematic reviews by offering a structured and data-driven ranking of relevant publications based on impact, recency, and citation scores. The results demonstrate a clear trend toward adopting ANN models in water treatment processes, with the multilayer perceptron neural network emerging as the most widely used model due to its high predictability and adaptability. Key validation criteria for ANN models are discussed, and recent experimental results related to their application in adsorptive systems for wastewater treatment are highlighted. This review not only underscores the potential of ANN tools for improving environmental control but also identifies key challenges and emerging trends in this rapidly evolving field.